Fourier Transforms Set To Become Faster @ SODA
@farjand-6UEF79
•
Oct 25, 2024
Oct 25, 2024
1.2K
Fourier transforms are one of the basic building blocks of information and technology sector. This is mainly because of the faster transactions that they ensure. Even so, we have algorithms more famous as Fast Fourier Transformations (FFT) which have made the data sending and receiving simpler. However, the question will always arise if it is possible to design faster algorithms than FFT. The answer is about to be discussed in Computing Machinery’s Symposium on Discrete Algorithms (SODA).
#-Link-Snipped-#
#-Link-Snipped-# have actually come up with a new concept which talks about increasing the speed of FFTs ten times the present. The main areas of it applications are hence thought to be image compression and video sharing. The solution thought out by the team is quite interesting as they claim to boost speed at little or no loss of signal quality. MIT's <a href="https://www.csail.mit.edu/" target="_blank" rel="nofollow noopener noreferrer">Home Page | MIT CSAIL</a> associate professor Dina Katabi and professor Piotr Indyk along with their students have developed an algorithm which appears to be an advanced version of working with Fourier transforms which we are seeing today.
The key factor on which the algorithm works is to express a digital signal as a weighted sum of equivalent number of frequencies. In other words, those frequencies which are present in less number are called low weight and are usually eliminated. This affects quality of file being processed a bit. However, the image remains intact. This will be used where you require speed more than any other feature. MIT's research group is hence trying to work on this Filtering action which is commonly seen in signal processing so that even the signal quality can be improved.
With the present research, the day is not far away when it will be ordinary to transfer big sized video files with negligible loss of its battery. Until now, this is not the only research in this regard. Let us wait for SODA for further updates!
#-Link-Snipped-#
#-Link-Snipped-# have actually come up with a new concept which talks about increasing the speed of FFTs ten times the present. The main areas of it applications are hence thought to be image compression and video sharing. The solution thought out by the team is quite interesting as they claim to boost speed at little or no loss of signal quality. MIT's <a href="https://www.csail.mit.edu/" target="_blank" rel="nofollow noopener noreferrer">Home Page | MIT CSAIL</a> associate professor Dina Katabi and professor Piotr Indyk along with their students have developed an algorithm which appears to be an advanced version of working with Fourier transforms which we are seeing today.
The key factor on which the algorithm works is to express a digital signal as a weighted sum of equivalent number of frequencies. In other words, those frequencies which are present in less number are called low weight and are usually eliminated. This affects quality of file being processed a bit. However, the image remains intact. This will be used where you require speed more than any other feature. MIT's research group is hence trying to work on this Filtering action which is commonly seen in signal processing so that even the signal quality can be improved.
With the present research, the day is not far away when it will be ordinary to transfer big sized video files with negligible loss of its battery. Until now, this is not the only research in this regard. Let us wait for SODA for further updates!